rfm = (
    df.groupby("customer_id")
    .agg(
        {
            "order_date": lambda x: (x.max() - x.min()).days,
            "order_id": "nunique",
            "price": "mean",
        }
    )
    .rename(
        columns={"order_date": "recency", "order_id": "frequency", "price": "monetary"}
    )
)

rfm_quantiles = rfm.quantile(q=[0.2, 0.4, 0.6, 0.8])

recency_conditions = [
    rfm.recency >= rfm_quantiles.recency.iloc[3],
    rfm.recency >= rfm_quantiles.recency.iloc[2],
    rfm.recency >= rfm_quantiles.recency.iloc[1],
    rfm.recency >= rfm_quantiles.recency.iloc[0],
    rfm.recency <= rfm_quantiles.recency.iloc[0],
]

frequency_conditions = [
    rfm.frequency <= rfm_quantiles.frequency.iloc[0],
    rfm.frequency <= rfm_quantiles.frequency.iloc[1],
    rfm.frequency <= rfm_quantiles.frequency.iloc[2],
    rfm.frequency <= rfm_quantiles.frequency.iloc[3],
    rfm.frequency >= rfm_quantiles.frequency.iloc[3],
]

monetary_conditions = [
    rfm.monetary <= rfm_quantiles.monetary.iloc[0],
    rfm.monetary <= rfm_quantiles.monetary.iloc[1],
    rfm.monetary <= rfm_quantiles.monetary.iloc[2],
    rfm.monetary <= rfm_quantiles.monetary.iloc[3],
    rfm.monetary >= rfm_quantiles.monetary.iloc[3],
]

ranks = [1, 2, 3, 4, 5]

rfm["r"] = np.select(recency_conditions, ranks, "other")
rfm["f"] = np.select(frequency_conditions, ranks, "other")
rfm["m"] = np.select(monetary_conditions, ranks, "other")

rfm["segment"] = rfm["r"].astype(str).add(rfm["f"].astype(str))

segment_map = {
    r"[1-2][1-2]": "hibernating",
    r"[1-2][3-4]": "at risk",
    r"[1-2]5": "cannot lose",
    r"3[1-2]": "about to sleep",
    r"33": "need attention",
    r"[3-4][4-5]": "loyal customers",
    r"41": "promising",
    r"51": "new customers",
    r"[4-5][2-3]": "potential loyalists",
    r"5[4-5]": "champions",
}

rfm["segment"] = rfm.segment.replace(segment_map, regex=True)